Vibrations in the Air surrounding a Sounding Body. 389 
the relations of delicate adjustments to be maintained durin 
each change of position, and therefore forks could only wit 
culty be made to serve in the measure of a wave-length, 
and could not at all be employed to trace out a wave-surface 
on account of the impossibility of a continuous comparison of 
their vibrations, which latter condition the manometric flames 
admirably fulfill. 
The Experiments. 
eh quite close to and directly behind, the organ pipe 
again caused the serrations of its flame neatly to bisect the 
Spaces between the serrations of the organ pipe flame, and moy- 
Ing around the o ipe, with the resonator held at such 
distances from it that the bisections were steadily kept, I de- 
Scribed in space the wave surface of the sounding pipe i 
Surface I found approximately to be an ellipsoid with its foci 
